5 papers
Planarian: Managing Agent State with Statepoints
Jinnan Guo, Hao Mark Chen, Kapil Vaswani +2
LLM agents solve complex tasks by iteratively changing files, invoking local tools, and interacting with remote services, which modifies state across their local environment and re…
Beyond Predictable Paths: Redefining AI Security Incident Reporting for Agents
Anastasia Pustozerova, Eugene Bagdasarian, Luca Beurer-Kellner +22
AI agents are being deployed rapidly, accompanied by a growing number of AI-specific attacks and corresponding incidents. As incident reporting becomes increasingly important for l…
Rising From the Ashes: How Agentic AI is Unblocking Challenges in Cybersecurity
Gabriela F. Ciocarlie, Kathrin Grosse, Somesh Jha +3
Security remains a high-cost challenge, with many problems historically deemed inefficient to address or effectively unsolvable. A significant number of these problems stem from la…
Practice-Informed, Practice-Ready: An AI security incident taxonomy
Lukas Bieringer, Sean McGregor, Nicole Nichols +7
With the increasing prevalence of AI systems, several real-world AI security incidents have been reported. However, despite forthcoming legal mandates, the reporting and collection…
VerifiableFL: Verifiable Claims for Federated Learning using Exclaves
Jinnan Guo, Kapil Vaswani, Andrew Paverd +1
In federated learning (FL), data providers jointly train a machine learning model without sharing their training data. This makes it challenging to provide verifiable claims about…